fix(mistral): preserve extended usage fields from API response#5693
Open
felipebridge wants to merge 1 commit into
Open
fix(mistral): preserve extended usage fields from API response#5693felipebridge wants to merge 1 commit into
felipebridge wants to merge 1 commit into
Conversation
The previous implementation only extracted prompt_tokens and completion_tokens, silently discarding any additional fields the Mistral API returns (e.g. cached_tokens, reasoning_tokens). It also used `if response.usage:` which evaluates truthiness on a model instance — always True even when all counts are zero. Use model_dump(exclude_none=True) to capture all fields, filter known token keys, and store the remainder in RequestUsage.details, matching the pattern used by the Groq provider. Fixes pydantic#5669
github-actions
Bot
added
size: S
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
_map_usageonly extractedprompt_tokensandcompletion_tokens, silently dropping any additional fields the Mistral API returns (e.g.cached_tokens,reasoning_tokens). It also had a latent bug:if response.usage:evaluates truthiness on a Pydantic model instance, which is alwaysTrueeven when all token counts are zero.This matches the pattern used by the Groq provider: call
model_dump(exclude_none=True)to capture all fields, filter the three known token keys, and forward any remaining integer fields toRequestUsage.details.Fixes #5669